- BFSIBanking, Financial Services & Insurance
- IT & ITESIT services, BPM, GCCs & SaaS
- Retail & e-commerceStore chains, D2C brands & marketplaces
- HealthcareComing soon
- ManufacturingComing soon
Databricks Unity Catalog
The auditor asks who can read the Aadhaar column. The answer is spread across six workspaces and forty views — Databricks Unity Catalog governs tables, files, models and AI services in one place — tag-driven access policies, column masks, automatic lineage and audit, with Unity Gateway for models and agents. Included with every Premium workspace.
Buy through TechBag
Same software. Better outcome — at a lower cost.
How it’s rated
Full scoreboard ↓Quick answer
The Databricks platform family
This page covers Databricks Unity Catalog — with Unity Gateway, Data Classification and Data Quality Monitoring folded in. The rest:
Quick facts
30-second orientation- Product
- Catalogue, access control, lineage and auditing for tables, files, models, functions and AI services
- Maker
- Databricks, Inc., San Francisco; also the creator of Apache Spark, Delta Lake and MLflow
- How it is bought
- Included in every Premium and Enterprise workspace; there is no separate Unity Catalog licence
- Open source
- Databricks lists Unity Catalog among the open-source projects it created, beside Spark and Delta Lake
- Folds in
- Unity Gateway (GA 4 Aug 2026), Data Classification, Data Quality Monitoring and Predictive Optimization
- Billed add-ons
- Managed services at the serverless jobs rate: $0.35 a DBU in US East, $0.37 in AWS Mumbai (Premium)
- Access model
- Attribute-based policies, row filters, column masks and governed tags; DENY policies are in Beta
- Analysts
- IDC MarketScape for Unified AI Governance Platforms, 2025: Leader; GigaOm Data Catalogs Radar, 2025: Leader
- India
- Runs wherever the workspace runs, including Mumbai and Pune; Data Classification there needs cross-Geo processing
- In India via
- TechBag — mapping access policies to your DPDP duties, a scoped rollout and an INR quote with GST
Understand data governance catalogues before you choose one
Most product pages skip this. We start here — so you buy a capability, not a buzzword.
What is Databricks Unity Catalog?
Unity Catalog is Databricks’ governance layer: one place for access, lineage and audit.
Per-workspace grants and masked view copies vs Databricks Unity Catalog — the honest table
What consolidation actually replaces, dimension by dimension.
| Dimension | Per-workspace grants and masked view copies | Databricks Unity Catalog |
|---|---|---|
| Access | Grants per workspace, copied by hand | One metastore, shared grants and tags |
| Sensitive data | Masked view copies of each table | Column masks and row filters on the source |
| Lineage | A diagram drawn once, then stale | Captured automatically as jobs run |
| AI services | API keys scattered across teams | Unity Gateway budgets and guardrails |
| Cost | A separate catalogue licence | Included; DBUs only for add-ons |
| What it is NOT | — | An estate-wide catalogue or a DSPM tool |
The cheapest test is one business domain: tag its sensitive columns, apply masks and row filters, then log in as each user group.
The four pieces of the platform
Vendors love diagrams; buyers need to know what they’re actually operating. Here’s the whole platform, demystified.
Metastore
One metastore per region holds catalogs, schemas, tables, volumes, models and functions, and every workspace attached to it shares the same permission model.
Policies
Role grants and attribute-based policies driven by governed tags decide access, while row filters and column masks hide records and fields from users who lack rights.
Lineage and audit
Lineage is captured automatically across tables and columns, external lineage is GA since June 2026, and a Discover page with domains (GA October 2026) helps users find data.
Unity Gateway
Unity Gateway, part of Unity Catalog and GA since 4 August 2026, routes model calls across providers and applies budgets, guardrails and usage tracking to agents and MCP servers.
A regional metastore holds every asset, tag-driven policies decide access, lineage and audit record what happened — and Unity Gateway extends the same control to AI.
Nine capabilities. Control, discover, monitor.
Databricks Unity Catalog puts data, models and AI services under one set of grants, tags and audit logs — included in your workspace.
Policies driven by tags
Attribute-based access control (GA) applies one policy to every column carrying a governed tag; cross-engine ABAC reached GA in October 2026.
Row filters and column masks
Row filters hide records and column masks redact fields such as PAN or Aadhaar numbers for users who lack the right to see them.
Access requests and secrets
Users request access in the catalogue (Public Preview), and Secrets in Unity Catalog (GA August 2026) put credentials under the same grants.
Lineage without set-up
Table and column lineage is recorded as queries and pipelines run, and external lineage from other tools has been GA since June 2026.
Sensitive columns found for you
Data Classification (GA) scans tables and tags columns that hold personal data, billed at the serverless jobs rate of $0.35 a DBU.
Domains and metric views
A Discover page organised by business domain (GA 8 October 2026) and governed metric views help users find trusted data and definitions.
Data Quality Monitoring
Data Quality Monitoring, formerly Lakehouse Monitoring, watches tables for drift and freshness; a 50% promotion runs until 31 January 2027.
Unity Gateway over AI usage
Unity Gateway sets which models and MCP servers teams may use, with budgets and guardrails that block or redact sensitive data in prompts.
Predictive Optimization
Predictive Optimization runs table maintenance such as compaction automatically on managed tables, billed in DBUs as a managed service.
Watch Databricks Unity Catalog in action
Three videos from Databricks’ official channel, all from 2026: a step-by-step Unity Catalog demo, a talk on governing AI with Unity AI Gateway, and a governed-agent build.
Getting Started with Unity Catalog: A Step-by-Step Databricks Demo
A 10-minute step-by-step demo from March 2026 for teams getting started with Unity Catalog on Databricks.
Governing AI Strategy with Unity AI Gateway
A 10-minute video from August 2026 on governing an organisation’s AI strategy with Unity AI Gateway.
Demo: Building a Governed AI Agent with Unity AI Gateway
A 14-minute demo, from October 2026, of building a governed AI agent with Unity AI Gateway.
Want a live, India-context walkthrough for your environment?
Book a guided demo →Most governance gaps hide between tools. Databricks Unity Catalog puts data and AI under one set of rules.
Here’s what genuinely sets it apart — and exactly where it stops.
One permission model for data and AI
Most estates keep table grants in one tool, model access in another and AI-service keys in a third. Unity Catalog puts tables, files, models, functions and, through Unity Gateway, AI services and MCP servers under one set of grants, tags and audit logs, so a masking rule written once follows the column everywhere it is read.
Already paid for, if you run Databricks
Unity Catalog has no licence of its own: it comes with every Premium and Enterprise workspace, including lineage and cross-workspace governance. You pay DBUs only for the managed services you switch on, such as classification or quality monitoring, which makes it the cheapest governance layer to start with on Databricks.
Open source and open formats
Databricks lists Unity Catalog among its open-source projects, and governed tables stay in open Delta Lake or Apache Iceberg formats in your own storage. GigaOm’s 2025 Data Catalogs Radar rated Databricks a Leader and Outperformer, and IDC’s 2025 MarketScape for unified AI governance a Leader.
Where it stops
Unity Catalog governs Databricks plus federated and open-format data; it is not built to catalogue every system you own the way Collibra or Purview aim to. Several features are still Beta or Preview, including DENY policies and access requests. Managed services add DBUs, and Data Quality Monitoring’s usage doubles after 31 January 2027.
The numbers behind the platform
What your Databricks Unity Catalog rollout looks like
Inventory and tag plan
List catalogs, sensitive columns such as PAN and Aadhaar, and the governed tags your DPDP and audit duties need.
Metastore and groups
Attach workspaces to a regional metastore in Mumbai or Pune, map identity groups and agree who owns each catalog.
Policies on one domain
Apply ABAC policies, column masks and row filters to one business domain, then test access as each user group.
Lineage, quality, audit
Check lineage for key reports, switch on classification and quality monitoring, and export audit logs to your SIEM.
Extend to AI and commit
Put model and MCP access behind Unity Gateway, roll policies out to other domains, then budget the add-on DBUs in INR.
The review scoreboard
Modelled on Gartner Peer Insights structure. *Counts and breakdowns are illustrative pending verified review collection.
Quick poll — what’s driving your evaluation?
Talk to an advisor“Column masks on PAN and Aadhaar fields replaced a dozen masked view copies, and the auditors liked one policy per tag.”
“Lineage came for free as pipelines ran, so impact analysis before a schema change now takes minutes instead of a week.”
“Unity Gateway let us cap each team’s model spend and see which agents call which MCP servers, all in one place.”
“It governs our Databricks estate well, but our SAP and on-prem Oracle still sit in a separate catalogue for stewards.”
“Classification tagged patient identifiers we had missed, though in India it needs cross-Geo processing, which we had to clear.”
“No licence to buy made the business case easy; the quality monitoring DBUs were the only line we had to budget for.”
Where everyone sits — the grids
Analyst firms bury this view behind paywalls, and review-site grids score satisfaction, not fit for your estate. So here’s TechBag’s synthesis of the data governance and catalogue market — tap any vendor to see why it sits where it does.
TechBag Data Governance Grid
Execution strength vs product vision — the classic market map, minus the paywall.
Included; IDC and GigaOm Leader (2025).
Enforcement Depth × System Reach
The grid nobody publishes — how far a catalogue reaches across the systems you own vs how deeply it enforces access itself, with masks, row filters and AI controls.
Deep enforcement and AI gateway; Databricks-first scope.
Positions are TechBag’s illustrative synthesis of public review-platform data and vendor documentation — not a reproduction of any analyst graphic. Verify before relying on it.
Databricks Unity Catalog vs the data-governance field
Against Snowflake Horizon Catalog, Microsoft Purview Unified Catalog, Google Knowledge Catalog, Amazon SageMaker Catalog and Collibra — on scope, access control, lineage, quality, AI governance, price and India.
| Dimension | Databricks Unity Catalog | Snowflake Horizon Catalog | Microsoft Purview Unified Catalog | Google Knowledge Catalog | Amazon SageMaker Catalog | Collibra |
|---|---|---|---|---|---|---|
| What it is | Databricks governance | Snowflake’s built-in layer | Microsoft data catalogue | Google Cloud catalogue | AWS data and ML catalogue | Independent catalogue |
| How it is bought | Included, not sold alone | Inside platform credits | Per asset, per day | Per DCU-hour | Free tier, then usage | Quoted subscription |
| Systems covered | Databricks plus federated | Snowflake plus Iceberg | Estate-wide scanning | Google data first | AWS data first | Many platforms |
| Fine-grained access | ABAC, rows, columns | Masking, row policies | Depends on source | IAM and policy tags | Subscribe and approve | Pushed to sources |
| Lineage | Automatic lineage | Within Snowflake | Varies by connector | Premium tier | AWS services | Core strength |
| Sensitive-data discovery | Data Classification | In-account classification | Classifies on scan | Profiling, separate DLP | Via Amazon Macie | Check the module |
| Data quality | Quality Monitoring | Data metric functions | Rules and scores | Quality scans | Via Glue | Quality product line |
| AI governance | Unity Gateway | Cortex AI Gateway | AI use oversight | Vertex AI side | Models in Studio | AI use-case registry |
| Openness | Open-source project | Polaris for Iceberg | Proprietary | Managed service | Managed service | Proprietary SaaS |
| Pricing model | Free with tier | Edition credits | Assets plus DGPUs | DCU-hours by tier | Requests and metadata | Quote or Marketplace |
| Published entry price | $0; add-ons $0.35/DBU | No separate price | $0.0165 per asset-day | 100 DCU-hours free | Free allowance | $170,000 a year |
| India data region | Follows the workspace | Mumbai and Pune | Central India | Check per region | Mumbai supported | Not published |
| Lock-in and exit | Open core, DBX policies | Snowflake-defined | Rebuild in a new tool | Google-specific | AWS-specific | Neutral, but sticky |
| Best fit | Databricks estates | Snowflake estates | Microsoft estates | Google Cloud shops | AWS lakehouses | Multi-platform enterprises |
Which approach fits you?
Honest fit signals — because the fastest way to lose your trust is to pretend one product wins every scenario.
Choose Databricks Unity Catalog if…
- ✓Databricks is where your data and AI work happens, and you want one set of grants for tables, models and agents
- ✓You want lineage, masking and AI-gateway controls without buying a separate catalogue licence
- ✓Open formats and an open-source catalogue matter to your exit plan
Compare alternatives if…
- ✓You need one catalogue for every system, SAP and on-premises included — Collibra or Purview aim at that
- ✓Your warehouse is Snowflake, BigQuery or Redshift — each cloud’s own catalogue is the natural fit
- ✓Stewardship workflows matter more than enforcement — Collibra is built around them
Do not expect…
- ✓A product you can buy on its own — it comes with Premium and Enterprise workspaces
- ✓A data-security posture tool — it governs access, it does not hunt exposures across clouds
- ✓Every feature to be GA — DENY policies, access requests and Governance Hub are still in preview
TechBag has no data governance and catalogues guide yet, so Databricks Unity Catalog sits outside the category guides. Browse all products to compare it with the rest of the catalogue. →
What does governing data by hand cost you?
Drag the sliders (tables holding sensitive data; data-engineer hour cost). Estimates model the time spent maintaining masked view copies, re-granting access per workspace and tracing lineage by hand for audits, at an assumed 1 hour per sensitive table a month, with 60% of it removed by tag-driven policies and automatic lineage. Both figures are assumptions. Illustrative.
Loaded cost = salary + overheads per productive hour. Illustrative only — your TechBag quote models your actual environment and modules.
Three ways to consume it
No separate price: Unity Catalog comes with every Premium and Enterprise workspace. Billed managed services run at the serverless jobs rate, USD list: $0.35 a DBU in US East and $0.37 in AWS Mumbai on Premium, $0.45 and $0.47 on Enterprise — Data Classification, Data Quality Monitoring (50% off until 31 January 2027), Predictive Optimization and fine-grained access on single-user clusters. Unity Gateway charges only for Usage Tracking and Inference Tables. TechBag quotes in INR with GST.
Unity Catalog
Best as the default for any Databricks estate
- Included in Premium and Enterprise
- Grants, ABAC, masks, lineage, audit
- No separate licence to buy
+ Platform add-ons
Best for a broader rollout
- Scoped to your estate
- Add-on modules as needed
- Phased, right-sized deployment
Managed services
Best once policies are in place
- $0.35/DBU US East, $0.37 AWS Mumbai
- Classification, quality, optimisation
- Quality Monitoring 50% off to Jan 2027
Buy it for less — TechBag pricing beats list
Whatever the list prices above, TechBag negotiates a significantly better deal — with GST-compliant INR invoicing and local support. Ask us for your discounted quote.
Get an India-ready quote
Tell us your requirements and current tools — we’ll model it against what you spend today.
The 8 questions to ask every vendor
Take this into your next vendor call — including ours.
Are all workspaces on Premium or Enterprise, since Unity Catalog is not available as a stand-alone purchase?
Which systems outside Databricks must be catalogued, and will federation cover them or is a second tool needed?
Which columns hold PAN, Aadhaar or health data, and which masks or row filters will DPDP duties require?
Is cross-Geo processing acceptable for Data Classification, or will you tag sensitive columns another way?
Where are audit logs exported, and do you keep them for at least the 180 days CERT-In’s directions require?
Do you rely on any Beta or Preview feature, such as DENY policies or access requests, for a control you must prove?
Which managed services will you switch on, and what will classification and quality monitoring cost each month?
What committed-use discount covers the add-on DBUs, and can TechBag itemise the quote in INR with GST?
Questions buyers ask
Ready to evaluate Databricks Unity Catalog?
List your sensitive data and the audits you face, or let a TechBag advisor map them to tags, masks and lineage, plan the metastore in Mumbai or Pune and itemise the add-on DBUs in INR.
Stats, ratings, review counts and pricing are illustrative and sourced from public materials; verify before purchase.